Bird Migration Monitoring in the Saint Nikola Wind Farm, Kaliakra Region, in Autumn 2016, and an Analysis of Potential Impact After Seven Years of Operation

Bird migration monitoring in the Saint Nikola Wind Farm, Kaliakra region, in autumn 2016, and an analysis of potential impact after seven years of operation

Dr. Pavel Zehtindjiev Institute of Biodiversity and Ecosystem Research, Bulgarian Academy of Sciences, 1113, Sofia, 2 Gagarin St., Bulgaria e-mail: [email protected]

Dr. D. Philip Whitfield Natural Research Ltd, Banchory, UK

November 2016

Report to AES Geo Energy OOD, 32A Cherni Vrah blvd, Sofia, Bulgaria

1 TERMS OF USE

You understand and agree that the information in, or derived from, this document may not be copied, republished, redistributed, transmitted, altered, edited, used or exploited in any manner for any purpose, without the express written permission of AES Geo Energy OOD ("AES"). You also agree that AES and its data providers shall not be liable for any errors in the content, or for any actions taken by you, or any third-party, in reliance thereon. Facts and other information discussed in this document have been obtained from sources considered reliable, but are not guaranteed, and AES makes no representation or warranty as to the accuracy or completeness of the information contained in this document or any other document or website referred to it or accessed through a hyperlink on AES' website. When you access a non-AES website, you understand that it is independent from AES, and that AES has no control over the content on that website. In addition, a link to a non-AES website does not mean that AES endorses or accepts any responsibility for the content, or the use, of such website. In no event will AES be liable to any party for any direct, indirect, special or other consequential damages for any use of this document, including, without limitation, any breach of law, any lost profits, business interruption, loss of programs or other data on your information handling system or otherwise, even if we are expressly advised of the possibility of such damages. All information is provided by AES on an "as is" basis only. AES provides no representations and warranties, express or implied, including the implied warranties of fitness for a particular purpose, merchantability and non-infringement. Except as explicitly stated otherwise, any notices of any dispute with respect to these Terms of Use or document shall be given by mail to AES Geo Energy OOD, 32A Cherni Vrah blvd, Sofia, Bulgaria. Any disputes arising out of your use of this document shall be governed in all respects by the laws of Bulgaria. Both parties submit to the jurisdiction of the Court of Arbitration at the Bulgarian Chamber of Commerce and Industry in compliance with its rules for litigation based on arbitration agreements.

2 Contents

SUMMARY ...... 4

INTRODUCTION ...... 4

METHODS ...... 5

THE STUDY AREA...... 5

STUDY DURATION AND EQUIPMENT ...... 5

BASIC VISUAL OBSERVATION PROTOCOL ...... 6

METHOD OF COLLISION VICTIM MONITORING ...... 8

STATISTICAL METHODS ...... 8

TURBINE SHUTDOWN SYSTEM (TSS) ...... 8

RESULTS AND DISCUSSION ...... 9

COMPOSITION OF SPECIES AND NUMBER OF BIRDS PASSING THROUGH SNWF ...... 9

ALTITUDE OF AUTUMN MIGRATION ...... 16

DIRECTION OF AUTUMN BIRD MIGRATION ...... 18

SPATIAL AND TEMPORAL DISTRIBUTION OF OBSERVED ‘MAJOR’ INFLUXES OF SOARING MIGRANTS AND TURBINE SHUTDOWN SYSTEM ...... 24

COLLISION VICTIM MONITORING...... 29

CONCLUSIONS ...... 32

REFERENCES ...... 33

3 SUMMARY

1. This report presents the results of 90 consecutive days of monitoring and mitigation at Saint Nikola Wind Farm (SNWF) in 2016, its 7th operational year. The continued purpose is to investigate the possible impacts of SNWF on migrating birds.

2. Spatial and temporal dynamics in the numbers of different species passing through the wind farm territory during autumn migration 2016 (15 August to 31 October) are presented. The data from the autumn monitoring in the years 2008 to 2016 are used to investigate the potential change in species composition, numbers, altitude or the flight direction of birds observed in these nine years at SNWF.

3. The variations in numbers of species, absolute number of birds, overall altitudes of flight and migratory direction of birds most sensitive to wind turbines do not indicate an adverse effect of the wind farm on diurnal migrating birds.

4. The Turbine Shutdown System (TSS) probably contributed to a reduced risk of collision during all years of operation within infrequent periods of intensive soaring bird migration and provided a safety mechanism to reduce collision risk for single birds and flocks of endangered bird species.

5. Regular searches under operational turbines for collision victims were continued, as in several previous years. In autumn 2016 these searches recorded only single casualties, for several species of no conservation concern: Magpie, Jay, Spotted flycatcher, Red- backed shrike, House martin, Kestrel, Goldcrest, Starling and Yellow-legged gull.

6. The predicted mortality rates by species based on preconstruction data on numbers of migrating birds are not supported by the mortality observed during any of the seven years of operation of SNWF. The levels of mortality predicted pre-construction have not been recorded during any year of operation. This is largely because ‘worst case’ predictions were based on BSPB (Bulgarian BirdLife partner) data that substantially exaggerated the numbers of migrants passing through SNWF.

7. The results to date continue to indicate that SNWF does not constitute a significant displacement/disturbing obstacle or mortality threat, either physically or demographically, to any of the populations of diurnal autumn migrants observed in this study.

INTRODUCTION

AES Geo Energy OOD constructed a 156 MW wind farm consisting of 52 turbines: the St Nikola Wind Farm (SNWF). In autumn 2008, SNWF did not exist; in autumn 2009 the facility was built but not operational (i.e. turbine blades were stationary), and in the autumns of 2010 - 2016 SNWF was operational.

In previous SNWF autumn reports the major focus was assessment of potential barrier effect on birds migrating through the territory and the level of collision mortality of migrants. The analysis of the data until now showed no evidence for cumulative long term changes in the migratory bird fauna. The main results of the autumn monitoring of bird migration in the vicinity of SNWF in previous years are published at: http://www.aesgeoenergy.com/site/Studies.html. In these studies negligible collision mortality

4 of migrating birds was found; indicating a high micro avoidance rate of the turbines by migrating bird species.

The present report updates the information on spatial distribution and temporal presence of birds in SNWF during autumn 2016 with, as in previous reports, special focus on soaring species deemed most sensitive to wind turbines. The observed increase of birds in SNWF in previous autumn seasons under westerly winds was tested statistically in a detailed correlative analysis of wind direction and bird numbers in autumn 2016.

Figure 1. Schematic representation of the main autumnal migratory flyway (blue arrows) and the location of SNWF (in red).

METHODS

The study area

SNWF is located in NE Bulgaria, approximately three to seven kilometers inland of the Black Sea coast and the cape of Kaliakra (Fig. 1). The wind farm lies between the road from the village of Bulgarevo to St. Nikola (municipality of Kavarna), and the 1st class road E 87 Kavarna – Shabla. The location of observation points is presented in Fig.2.

Study duration and equipment

The study was carried out between 15 August and 31 October 2016 using standard methods that are comparable for all nine autumn seasons since studies began in 2008, using up to six field ornithologists making visual observations. The surveys were made as in previous seasons during the day, in a standard interval of time between 8 AM and 6 PM astronomic time (for details see http://www.aesgeoenergy.com/site/Studies.html.)

5 Basic Visual Observation Protocol

The autumn 2016 study involved direct visual survey of all passing birds from several observation points (Fig. 2). Field observations followed the census techniques according to Bibby et al. (1992). Point counts were performed by scanning the sky in all directions. Height estimates and distances to the birds were verified with land mark constructions around the observation points previously measured and calibrated by GPS. The surveys were carried out by means of optics, every surveyor having a pair of 10x binoculars and all observation points were equipped with 20 – 60x telescope, compass, GPS, and digital camera.

Figure 2. Map of the "SNWF" study area (red plot), and the "core study area" (brown area) covered by the autumn monitoring 2016 observations and location of the observation points (white circles).

As noted in previous reports, 2009 was exceptional in the spatial survey protocol because the observation points were moved northward to test the early warning system (TSS) for approaching flocks of birds. The northerly shift in the observation points in 2009 means that many data of migratory metrics (notably, flight direction) were likely not comparable with the years before or since. In 2009, SNWF had been constructed but was not operational. The basic temporal survey protocol was otherwise not changed in the period 2008 – 2016 (other than the temporal extension in 2013 to 2016 to cover October, additionally) in order to allow comparable data collection between years.

As described in several previous reports, it was apparent in earlier years that the occurrence of relatively unusual westerly winds was the main reason for influxes of soaring birds in SNWF territory. Hence this feature has been subjected to detailed analysis in this autumn 2016 report.

All details about the specific visual observation protocol are presented in a number of previous autumn reports and in the Owner Monitoring Plan (OMP) and will not be repeated here: http://www.aesgeoenergy.com/site/images/21.pdf (studies page).

6 All observers were qualified specialists in carrying out the surveys of bird migration for many years including previous autumn surveys at SNWF.

List of participants in the autumn observations, 2016 Dr Pavel Zehtindjiev - Senior Field Ornithologist Institute of Biodiversity and Ecosystem Research Bulgarian Academy of Sciences Victor Metodiev Vasilev - Field ornithologist Senior researcher in the Faculty of Biology University of Shumen, Bulgaria BSPB (Bird Life Bulgaria) member Ivailo Antonov Raykov - Field ornithologist Museum of Natural History, Varna BSPB (Bird Life Bulgaria) member Strahil Georgiev Peev - Field ornithologist Qualified carcass searcher PhD Student, Institute of Biodiversity and Ecosystem Research BSPB (Bird Life Bulgaria) member Kiril Ivanov Bedev - Field ornithologist Qualified carcass searcher Institute of Biodiversity and Ecosystem Research Yanko Sabev Yankov - Field ornithologist Qualified carcass searcher Student in Biology BSPB (Bird Life Bulgaria) member Martin Petrov Marinov - Field ornithologist Qualified carcass searcher PhD Student, Institute of Biodiversity and Ecosystem Research Bulgarian Academy of Sciences Karina Ivailova Ivanova - Field ornithologist PhD Student, Institute of Biodiversity and Ecosystem Research Bulgarian Academy of Sciences

Nikolai Sashov Bunkov - Field ornithologist PhD Student, Institute of Biodiversity and Ecosystem Research Bulgarian Academy of Sciences

As already stated, over the years 2008-2012 the autumn monitoring lasted for the period of most intensive migration - August and September. Since 2013 (including 2016), we have extended the period of observation until the end of October. In order to provide comparability between the four most recent seasons and previous years, however, to avoid bias associated with the extended observation period in 2013 to 2016, the data presented below are based on a comparable time period (15 August to 30 September) unless otherwise stated.

7 Method of Collision Victim Monitoring

The collision monitoring methodology followed that developed in the USA for bird collision monitoring at wind farms (Morrison 1998). The detailed description of the protocol is given in par. 1.6 and 2.4 of the Owners Monitoring Plan (OMP http://www.aesgeoenergy.com/site/Studies.html.). Staged autumn trials were conducted in two previous years examining carcass removal/disappearance rates and searcher efficiency rates. These results, presented in previous autumn reports, should be borne in mind as adjustment factors when considering the results for carcasses and numbers found during the systematic searches under turbines during 2016.

Statistical methods

The number of observed species, individuals as well as their average altitude of flight (by species and years) is presented in a number of tables for direct comparison across the autumn seasons of 2008 - 2016.

The altitude of migration in different autumn seasons was evaluated for significance by its mean value, standard error and standard deviation in data analysis software system STATISTICA (StatSoft, Inc. (2004, version 7. http://www.statsoft.com/). The mean flight direction as well as its significance level, for every species and group of species was calculated according to standard circular statistics (Batschelet 1981). Circular statistics was performed with Oriana (Oriana - Copyright © 1994-2009 Kovach Computing Services). This program compares two or more sets of circular distributions (directions) to determine if they differ. The tests were performed pairwise, so that each pair of samples was compared separately.

Many of the basic statistical parameters of circular distributions (directions) are based on the concept of the mean vector. A group of observations (or individual vectors) have a mean vector that can be calculated by combining each of the individual vectors (the calculations are explained in most books about circular statistics). The mean vector has two properties; its direction (the mean angle, µ) and its length (often referred to as r). The length ranges from 0 to 1; a higher r value indicates that the observations are clustered more closely around the mean than a lower one. Details about the Oriana software are available at: http://www.kovcomp.com/ Wind direction was recorded by a permanent meteorological station set up at SNWF. A correlation between predominant prevailing daily wind direction and number of birds recorded daily was performed using the software Statistica 8 for Windows (StatSoft, Tulsa, OK, USA).

Turbine Shutdown System (TSS)

The principles to selectively stop specific turbines or the entire wind park to reduce risk of collisions are described in par. 1.5 of the Owners Monitoring Plan (OMP).

The TSS protocol was followed in order to reduce collision risk during the extended period of study in autumn 2016, between 15 August and 31 October. Turbine shutdowns are ordered by the Senior Field Ornithologist or - when delegated - to field ornithologists in the case of any perceived collision risk to an influx of potentially collision-sensitive species.

8 RESULTS AND DISCUSSION

Composition of species and number of birds passing through SNWF

The occurrence of species across all years is presented in Table 1. A total of 128 bird species have been observed in the wind farm territory during the consecutive autumn seasons of 2008 to 2016. The number of observed species varied from 48 to 82 in different years. 33 species were observed every autumn season in the period 2008 – 2016. Regular migrants through the territory included White Pelican, White Stork, Levant Sparrowhawk, Common Buzzard, Honey Buzzard and the Lesser Spotted Eagle.

By contrast, another 52 species of birds were not recorded in 2008, but observed at least in one of eight post-construction autumn seasons. Among such species were, for example, many birds of prey like Golden Eagle, Saker Falcon, Black Kite; waders like Northern Lapwing, Green Sandpiper, Common Greenshank, Eurasian Stone-curlew; herons like Purple Heron, Great Egret, Little Egret; and many small passerine bird species. The occurrence of these relatively rare species after construction should be attributed to vagrancy. Three new species observed in autumn 2016 are Steppe Eagle (Aquila nipalensis), Rough-legged Buzzard (Buteo lagopus) and a flock of 41White-fronted geese. The steppe Eagle is a rare for Europe and cannot be typically associated with autumn migration in the region. Rough-legged Buzzard and White fronted geese are wintering in the area of SNWF, but their appearance in autumn is observed for the first time for nine years of our monitoring. There is no apparent substantive difference in composition of species migrating through the wind farm observed in 2008 (before the construction of the wind farm) and during the later period when the wind farm was present (2009 – 2016). No species recorded in 2008, before SNWF was constructed, has not been recorded subsequently in years after construction; and several species have been recorded in the seven years after construction that were not recorded in 2008. While this can illustrate that SNWF has not impaired the occurrence of species on migration, such differences should not be attributed to any ‘beneficial’ effects of SNWF but to the greater number of years of observation post-construction. Table 1. List of species observed in SNWF during 15 August to 30 September in pre-construction (2008) and post-construction (2009 to 2016 in grey) periods of SNWF. Hatched cells represent the years when the species was registered in SNWF. N Species 2008 2009 2010 2011 2012 2013 2014 2015 2016 1 A. albifrons 2 A. apus 3 A. arvensis 4 A. brevipes 5 A. campestris 6 A. cervinus 7 A. chrysaetos 8 A. cinerea 9 A. gentilis 10 A. heliaca 11 A.nipalensis 12 A. melba 13 A. nisus 14 A. pennata 15 A. pomarina

9 N Species 2008 2009 2010 2011 2012 2013 2014 2015 2016 16 A. pratensis 17 A. purpurea 18 A.rapax 19 A. trivialis 20 B. buteo 21 B. oedicnemus 22 B. rufinus 23 B.b. vulpinus 24 B.lagopus 25 C. aeruginosus 26 C. cannabina 27 C. canorus 28 C. carduelis 29 C. chloris 30 C. ciconia 31 C. coccothraustes 32 C. corax 33 C. cornix 34 C. coturnix 35 C. cyaneus 36 C. frugilegus 37 C. gallicus 38 C. garrulus 39 C. livia domestica 40 C. macrourus 41 C. monedula 42 C. nigra 43 C. olor 44 C. palumbus 45 C. oenans 46 C. pygargus 47 D. major 48 D.syriacus 49 D. urbica 50 E. alba 51 E. calandra 52 E. garzetta 53 E. hortulana 54 E. melanocephala 55 F. cherrug 56 F. coelebs 57 F. eleonorae 58 F. naumanni 59 F. parva 60 F. peregrinus 61 F. subbuteo 62 F. tinnunculus

10 N Species 2008 2009 2010 2011 2012 2013 2014 2015 2016 63 F. vespertinus 64 G. fulvus 65 G. glandarius 66 G. grus 67 G. cristata 68 H. daurica 69 H. icterina 70 H. pallida 71 H. rustica 72 H. albicilla 73 J. torquila 74 L. cachinnans 75 L. collurio 76 L. megarhynchos 77 L. melanocephalus 78 L. minor 79 L. ridibundus 80 M. alba 81 M. apiaster 82 M. calandra 83 M. cinerea 84 M. flava 85 M. migrans 86 M. milvus 87 M. striata 88 N. percnopterus 89 O. hispanica 90 O. isabellina 91 O. oenanthe 92 O. oriolus 93 O. pleschanka 94 P. apivorus 95 P. caeruleus 96 P. crispus 97 P. haliaetus 98 P. leucorodia 99 P. major 100 P. montanus 101 P. onocrotalus 102 P. perdix 103 P. pica 104 P. viridis 105 Ph. carbo 106 Ph. collybita 107 Ph. trochilus 108 Pl. falcinellus 109 Ph. pygmaeus

11 N Species 2008 2009 2010 2011 2012 2013 2014 2015 2016 110 Ph. ochrurus 111 Ph. phoenicurus 112 R. riparia 113 S. borin 114 S. communis 115 S. curruca 116 S. rubetra 117 S. vulgaris 118 St. hirundo 119 Str. decaocto 120 Str. turtur 121 T. nebularia 122 T. glareola 123 T. tadorna 124 T. ochropus 125 T. merula 126 T.viscivorus 127 U. epops 128 V. vanellus Number of species 77 82 48 71 79 81 79 66 60

The observed variations in the number of species observed in the study area is due to the vagaries of rare bird species’ occurrence which in any year are present in low numbers and therefore observed sporadically in some autumns: Common Crane, Griffon Vulture, Egyptian Vulture, Imperial Eagle, Golden Eagle, Red Kite, Saker Falcon, Lesser Kestrel and Eleonora's Falcon, Eagle, Dalmatian Pelican, and Lesser Kestrel.

Surprisingly a flock of 41 Greater White-fronted geese (Anser albifrons) was observed on 24 October, much earlier than this species usually appears in the wintering grounds. Another ‘new’ species observed in autumn 2016, Steppe Eagle breeds in Asia and Bulgaria is outside of its distribution. Appearance of Steppe Eagles in Bulgaria is usually considered as rare observation. The second ‘new’ species, Rough-legged Buzzard, breeds in northern latitudes, and winters in more southern parts of the Palearctic. Its appearance during autumn migration is too early with respect to typically winter arrivals of this species in Bulgaria. Two of the most sensitive species with respect to collision with turbines, according to the literature, are Griffon Vulture (Gyps fulvus) and Egyptian Vulture (Neophron percnopterus). Both species have been observed, albeit in small numbers, in autumn monitoring periods after SNWF construction (see previous reports). In 2016 one Griffon Vulture was observed on 28 September at 150 m altitude, crossing SNWF territory. Egyptian Vultures were not observed in autumn 2016.

Absolute counts of soaring species which were most numerous, together with some additional species with high conservation value, are presented in Table 2. Table 2. Numbers of birds recorded as passing through SNWF (primarily soaring water birds and birds of prey) in nine autumn seasons of pre-construction (2008) and post-construction years (2009 – 2016). Species 2008 2009 2010 2011 2012 2013 2014 2015 2016 A. brevipes 95 210 976 290 94 650 138 190 334

12 Species 2008 2009 2010 2011 2012 2013 2014 2015 2016 A. chrysaetos 2 2 1 1 2

A. cinerea 120 259 26 40 56 70 113 20 50 A. gentilis 10 6 5 11 22 38 9 16 4 A. heliaca 2 1 A. nisus 44 44 70 73 44 206 101 133 150 A. pennata 4 3 22 5 10 22 14 10 8 A. pomarina 44 9 80 76 31 1966 509 146 18 A. purpurea 59 11 1 7 3 2 B. buteo 146 390 180 459 238 2345 1073 499 856 B. oedicnemus 1 1

B. rufinus 163 151 34 30 33 28 41 32 27 C. aeruginosus 327 268 341 271 179 473 298 339 165 C. ciconia 2998 87 24980 620 2525 11230 4639 292 1191 C. cyaneus 5 1 1 3 18 3

C. gallicus 29 19 18 25 60 88 26 38 27 C. macrourus 8 27 18 4 7 7 15 8 2 C. nigra 8 8 8 1 13 488 48 29 25 C. olor 1 3 2 11 C. palumbus 10 1 26 2 C. pygargus 32 17 111 151 55 82 102 161 47 E. alba 1 1 5

E. garzetta 7 11 1 33 F. cherrug 7 2 1 1

F. eleonorae 7 1 1 7

F. naumanni 1

F. peregrinus 2 4 1 1 5 5 2 1 F. subbuteo 48 125 120 96 66 88 89 135 31 F. tinnunculus 138 357 45 120 67 103 89 108 86 F. vespertinus 11 180 1773 63 793 167 426 434 107 G. fulvus 1 1 2 1 1 1 G. grus 1 91 32 M. migrans 18 6 32 17 21 34 32 69 8 M. milvus 1 1 2 1 1 N. percnopterus 1 2 P. apivorus 58 76 1549 152 115 4284 113 258 55 P. crispus 4 5 21

P. haliaetus 15 13 14 12 7 13 5 20 13 P. leucorodia 117 83 56 48 59 122 22 P. onocrotalus 120 1190 252 277 1700 3285 1679 2857 1527 Ph. carbo 267 354 494 75 131 866 263 542 Ph. pygmaeus 19

Pl. falcinellus 5 738

St. hirundo 71

T. tadorna 94 3

T. ochropus 8 1 15

T. glareola 3 11 T. merula 80

T. viscivorus 17

13 Species 2008 2009 2010 2011 2012 2013 2014 2015 2016 V. vanellus 1 7 7 Total 4854 4890 31229 2927 6288 25761 10594 6332 5353 Number of species 30 35 32 32 31 31 36 34 28

The number of species as well as the absolute number of birds crossing the study area (Tables 1 and 2) did not decrease after the construction of turbines. The absolute number per year of the most numerous species of soaring migrants; White Pelican, White Stork, Levant Sparrowhawk, Common Buzzard, Honey Buzzard and Lesser Spotted Eagle, widely varied in the nine study seasons (Fig. 3 & 4).

Figure 3. Variations in the total number of the most numerous soaring bird species observed during autumn migrations in nine years (pre-construction 2008, and post-construction periods - in background grey shading) in SNWF.

Figure 4. Percentage annual contribution of individual species (of the six most numerous soaring bird species recorded) to the total migratory traffic in and over SNWF in autumns 2008 – 2016 (pre-construction 2008, and post-construction periods - in background grey shading).

Another numerous group of migrants recorded at SNWF are species specialized in diurnal aerial foraging for insects. Not all birds of these species, bee-eaters, swifts and swallows (hirundines), crossing SNWF were detected because of their small size and methodological limitations of visual observations. The recording of these species highly depends on the distance from the observer (in both vertical and horizontal visual planes) because of their small size and, often their flight altitude (for details see autumn report 2013). Therefore visual observations on these species are limited to a few hundred meters and cannot be considered as absolute numbers for a given area and at all altitudes.

In autumn 2016 the number of Swifts and swallows was obviously lower (Table 3). One possible explanation could be prevailing winds with eastern components which does not drift aerial foraging birds into the area of SNWF.

With these caveats in mind, the results on the numbers of bee-eaters and hirundines (swallows and swifts) (hirundines not identified to the species level are not presented) registered between 2008 and 2016 are given in Table 3.

Table 3. The number of bee-eaters, swifts and swallows in SNWF in nine autumn seasons as observed in the period 15 August – 30 September. Species 2008 2009 2010 2011 2012 2013 2014 2015 2016 A. apus 79 10 6 8 17 12 52 39 4 A. melba 515 16 536 234 47 127 58 26 8 D. urbica 1007 697 180 3 170 109 436 25

15 Species 2008 2009 2010 2011 2012 2013 2014 2015 2016 H. daurica 2 8 4 1

H. rustica 2979 4234 1735 164 5994 815 550 473 40 M. apiaster 4625 3355 5024 2107 2733 5906 1828 1377 688

Altitude of autumn migration

In order to test whether the construction of SNWF turbines has resulted in an increase of flight altitude of migrating birds we calculated the average altitude per year of all species of diurnal migrants regularly passing through SNWF in autumn, including 2016 (Table 4).

Table 4. Mean flight altitude (in meters above the ground level), by species, of diurnal migrants observed in SNWF across nine autumn seasons, 2008-2016: the years when the wind farm was constructed are highlighted in grey. 2008 2009 2010 2011 2012 2013 2014 2015 2016 Species

A. brevipes 132 171 171 160 142 263 188 178 175 A. cinerea 201 239 263 386 190 344 341 133 288 A. gentilis 181 176 230 199 151 267 232 146 65 A. nisus 150 135 162 141 119 204 124 139 170 A. pennata 150 283 251 213 295 261 368 213 255 A. pomarina 244 273 234 234 241 353 279 210 243 B. buteo 165 199 206 197 158 278 215 187 202 B. rufinus 109 200 230 183 147 211 177 156 165 C. aeruginosus 158 139 235 150 128 222 201 113 113 C. ciconia 199 174 434 347 358 390 279 242 296 C. cyaneus 136 100 10 267 70 100 11 C. gallicus 256 144 258 242 218 229 269 221 190 C. macrourus 251 90 240 195 86 188 150 98 53 C. nigra 462 325 375 350 388 382 330 339 260 C. pygargus 196 115 285 106 79 209 144 107 126 F. subbuteo 97 119 161 161 127 131 181 139 94 F. tinnunculus 49 96 109 70 79 67 85 40 55 F. vespertinus 106 106 224 289 121 139 156 197 226 M. migrans 175 183 166 152 233 243 179 213 236 P. apivorus 320 175 268 283 204 342 290 270 240 P. haliaetus 314 208 224 433 400 133 172 303 P. leucorodia 433 285 667 317 317 350 500 P. onocrotalus 100 159 417 400 265 263 271 230 275 Ph. carbo 180 179 277 271 254 265 285 284 285

No trend in the fluctuations of average altitude of the most numerous soaring bird species was registered after nine years of autumn migration monitoring at SNWF, including one pre- construction and eight post-construction seasons. The comparative analysis showed that there was no significant change in average flight altitudes of the 24 most numerous soaring bird species regularly migrating through SNWF (Fig. 5).

16 700

600

500

400

300

200

100 Median 25%-75% Non-Outlier Range Outliers 0 Extremes 2008 2009 2010 2011 2012 2013 2014 2015 2016

Figure 5. The median altitude of soaring bird migration observed from SNWF during autumns of 2008 to 2016, with measures of variance. The species included in the calculations are presented in Table 4.

Observed flight altitudes of bee-eaters and swallows were analyzed despite the constraints on reliability imposed by visual observation, as previously noted. Nevertheless, despite the caveats on observational constraints (which should apply more-or-less equally across study years), it appeared that while the average observed flight altitude of bee-eaters and swallows varied widely across years there was no trend that could be attributable to the presence of SNWF (Table 5).

Table 5. Mean altitude of flight during autumn migration of bee-eaters M. apiaster and barn swallows H. rustica in the period 2008 – 2016 observed in SNWF. Species 2008 2009 2010 2011 2012 2013 2014 2015 2016

H. rustica 28 51 66 19 37 32 35 35 50 M. apiaster 73 68 128 71 83 66 85 100 92

Changes in the flight altitude of soaring migrants, bee-eaters and swallows have apparently had no consistent character across years and do not indicate any impact due to SNWF. Most probably climatic factors, conditions on the breeding grounds of these species that breed away from SNWF, and local aerial insect availability at the time of passage (for those species in Table 5) are likely to be responsible for the fluctuations in average altitude of autumn

17 migration in the nine year monitoring period. Regardless, any energetic consequences for migrants avoiding the turbines by way of a change in flight altitude will be immaterial to overall migratory energy budgets (Madsen et al. 2009, 2010) if they occur.

Direction of autumn bird migration

The mean recorded direction of the 24 species (listed in Table 4) is presented in Table 6. Prevailing directions of autumn migration observed in all nine autumn seasons do not indicate changes in migratory direction through a response to SNWF in years when there was greater consistency in the location of observation points (i.e. excluding 2009 when the observation points were moved northward in order to test the TSS). The main direction in all years shows the guiding role of the coast line (see Fig. 1 and Table 7).

Table 6. Mean observed flight direction of autumn migration by species listed in Table 4, in different years. Directions are given in degrees starting from 0 (North). Species 2008 2009 2010 2011 2012 2013 2014 2015 2016 A. brevipes 172 151 185 175 179 191 156 161 166 A. cinerea 248 178 146 138 203 167 176 101 169 A. gentilis 195 162 171 180 149 181 163 188 90 A. nisus 218 155 186 193 174 185 164 164 174 A. pennata 180 150 182 165 216 184 212 198 128 A. pomarina 225 173 204 183 193 214 180 196 166 B. buteo 195 150 177 179 179 198 172 165 166 B. rufinus 150 158 227 186 188 158 119 185 169 C. aeruginosus 197 150 191 188 175 199 166 166 154 C. ciconia 207 154 209 210 209 216 181 215 206 C. cyaneus 90 180 225 188 180 135 135 C. gallicus 203 150 144 151 129 159 142 165 130 C. macrourus 141 154 180 231 109 210 144 135 203 C. nigra 270 191 225 180 231 205 163 206 180 C. pygargus 237 148 182 183 174 194 154 165 165 F. subbuteo 186 148 174 196 196 188 157 156 157 F. tinnunculus 144 148 177 161 191 156 153 138 175 F. vespertinus 180 159 177 204 218 206 169 198 186 M. migrans 241 153 211 207 189 192 210 179 203 P. apivorus 227 187 201 200 208 204 174 195 176 P. haliaetus 161 190 168 198 169 199 152 135 168 P. leucorodia 180 173 195 180 180 162 180 P. onocrotalus 146 195 257 232 214 180 177 15 Ph. carbo 178 162 192 160 121 177 155 154 132

Table 7. Basic statistical parameters of empirical flight directions obtained from visual observations during nine autumn seasons in SNWF for the 24 ‘core’ soaring bird species (listed in Table 4). Autumn season 2008 2009 2010 2011 2012 2013 2014 2015 2016 Number of species 23 24 23 24 22 24 23 24 24 Mean Vector (µ) 193° 161° 186° 188° 184° 190° 166° 168 164 Length of Mean Vector (r) 0,8 0,96 0,93 0,90 0,85 0,95 0,94 0,89 0,82 Concentration 2,7 16,6 8,4 5,5 3,7 11,8 8,8 5,1 3,2

18 Autumn season 2008 2009 2010 2011 2012 2013 2014 2015 2016 Circular Variance 0,21 0,03 0,06 0,09 0,14 0,95 0,05 0,1 0,17 Circular Standard 39,3° 14,2° 20,2° 25,5° 32,3° 17,1° 19,8° 26,6 35,4 Deviation

The circular (compass) distributions of flight directions of soaring birds are presented in graphs below for each year (Fig. 6).

Figure 6. Graphical representations of the average flight directions of the 24 ‘core’ soaring bird species by year: each record = 1 species (see Tables 4, 6 and 7). (In 2009, observation points were stationed further north than in other years.)

The direction of migration in 24 of the most common and numerous soaring birds observed at SNWF in the last nine years does not indicate any consistent annual deviation from the seasonal migratory direction after construction of SNWF (Table 7 and Fig. 6). An expectation, if the turbines were causing birds to avoid the study area would be that there should be a major shift in migratory direction much further to the west, as birds deflect inland and away from the wind farm. This has not been recorded.

In 2014, 2015 and 2016 the mean direction of the same most numerous species of soaring birds suggested that not only the location of observation points (as in 2009) but also some other factors (perhaps conspecific flock attraction and probably specific wind directions during the season) may also explain annual deviations from the typical direction of soaring bird migration across SNWF over the nine years of study.

Bearing in mind the feeding behavior of bee-eaters and swallows which are specialized in hunting insects in the air during daytime, and the detailed analysis of flight directions in previous reports, it is also likely that several species’ abundance may be governed by the capacity for feeding activity as well as active migratory flight through SNWF during autumn (Table 8).

Table 8. Mean flight directions of barn swallows H. rustica and bee-eaters M. apiaster as observed from SNWF across nine autumn seasons. Directions are given in degrees starting from 0 (North). Species 2008 2009 2010 2011 2012 2013 2014 2015 2016 H. rustica 158 144 204 169 172 150 101 68 Low number M. apiaster 191 142 192 186 187 189 177 162 151

23 There is no evidence under the scale and form of analysis for a major directional change in the flight orientation behavior of autumn migrants (macro-avoidance) as a result of the wind farm operation. At the scales considered, birds that were observed to enter the vicinity of the wind farm did not demonstrate any macro-avoidance of the turbines which could thereby be considered as a change of migratory direction and, consequently, contribute to a major change in migratory route or any detrimental effect on energy budgets.

Spatial and temporal distribution of observed ‘major’ influxes of soaring migrants and Turbine Shutdown System

In autumn 2016, intensive soaring bird migration was observed mainly in the standard monitoring period 15 August – 30 September defined in previous reports with a peak period in September (Fig. 7). Prevailing wind directions in autumn 2016 were N – NE (Table 9); the same as in every previous autumn of the study. Again as in previous years, westerly winds, which bring periodic influxes of soaring migrants swept easterly from the main Via Pontica migration route (Fig. 1) were infrequent.

Figure 7. Monthly distribution of all registrations of migrating birds during the autumn season 2016.

Notable days with relatively strong migration of soaring birds at low altitudes was observed on 3 August and 26 September with 700 White storks Ciconia ciconia and 500 White pelicans (Pelecanus onocrotalus) , respectively. Notable numbers were observed also on 6 October when 34 White pelicans (Pelecanus onocrotalus)crossed the SNWF territory. All the events of turbine stops in respond to target bird species presence in SNWF are listed in Table 10.

24 Table 9. Number of birds and wind direction diring the autumn 2016 monitoring period. For reference: a northerly wind direction = 0, and a southerly wind direction = 180. Date Number Wind Date Number Wind Date Number Wind of birds direction of birds direction of birds direction 1.8. 3 74 1.9. 9 144 1.10. 5 132 2.8. 3 133 2.9. 15 91 2.10. 12 97 3.8. 701 255 3.9. 10 154 3.10. 18 163 4.8. 26 129 4.9. 3 190 4.10. 9 268 5.8. 2 46 5.9. 5 179 5.10. 400 256 6.8. 4 75 6.9. 14 94 6.10. 134 209 7.8. 51 190 7.9. 9 36 7.10. 14 171 8.8. 2 137 8.9. 14 52 8.10. 4 273 9.8. 5 20 9.9. 10 128 9.10. 3 243 10.8. 3 49 10.9. 66 90 10.10. 3 172 11.8. 9 155 11.9. 105 58 11.10. 1 111 12.8. 10 322 12.9. 21 53 12.10. 8 220 13.8. 11 214 13.9. 56 172 13.10. 22 277 14.8. 9 226 14.9. 152 216 14.10. 20 334 15.8. 10 185 15.9. 70 159 15.10. 3 208 16.8. 13 179 16.9. 8 136 16.10. 1 77 17.8. 11 161 17.9. 33 199 17.10. 2 45 18.8. 24 175 18.9. 20 177 18.10. 1 59 19.8. 14 84 19.9. 8 153 19.10. 16 255 20.8. 4 91 20.9. 38 316 20.10. 7 114 21.8. 5 203 21.9. 79 274 21.10. 4 83 22.8. 8 166 22.9. 428 222 22.10. 2 99 23.8. 512 312 23.9. 11 318 23.10. 2 59 24.8. 13 284 24.9. 4 259 24.10. 3 34 25.8. 12 173 25.9. 13 207 25.10. 33 158 26.8. 6 178 26.9. 1010 190 26.10. 3 79 27.8. 13 193 27.9. 3 249 27.10. 2 331 28.8. 24 171 28.9. 21 249 28.10. 10 284 29.8. 17 117 29.9. 42 277 29.10. 9 277 30.8. 17 256 30.9. 6 208 30.10. 45 298 31.8. 15 184 31.10. 228 314

25 Table 10. List of observed ‘major’ influxes of soaring migrants according to species, in autumn 2016 in or over SNWF, by date and the stop and start times of turbine shutdowns.

Date Stop Start Species Number of Ordered by Wind direction the birds

03.08.16 14:11 14:20 White stork 700 S. Peev NW

23.08.16 11:31 11:35 Steppe eagle 1 S. Peev NW

18.09.16 10:11 10:35 White pelican 1 M. Marinov No wind

26.09.16 11:00 11:15 White pelican 500 V. Vaslev N

28.09.16 10:10 10:20 Griffon vulture 1 Y. Yankov NW

28.09.16 10:26 10:40 Griffon vulture 1 K.Bedev NW

28.09.16 10:30 11:32 Griffon vulture 1 K.Bedev NW

28.09.16 10:36 11:32 Griffon vulture 1 K.Bedev NW

03.10.16 12:37 12:43 White Pelican 1 Y. Yankov SE

06.10.16 11:31 11:36 White pelican 34 K.Bedev SE

06.1016. 11:36 11:47 White pelican 34 K.Bedev SE

06.10.16 11:55 12:47 White pelican 34 K.Bedev SE

Our long term monitoring of autumn migratin in SNFW has revealed an increase of birds in the days with western winds (see report autumn 2010 http://www.aesgeoenergy.com/site/images/Bird_Migration_autumn_2010.pdf ). In order to perform statistical tests and evaluate the significance of winds with a westerly direction for observed increase in soaring bird numbers we have applied a statistical test through a correlative analysis. Results are presented in Fig. 8 and 9.

26 Figure 8. Number of soaring birds (blue line) and wind direction day by day(red line) in SNWF in autumn 2016.

Relatively lower numbers of soaring birds were observed in SNWF in autumn 2016 (for species and number of birds, see Table 4) and were concentrated in six days when prevailing winds were with a strong westerly component (Fig. 8). The bird species included in this analysis are presented in Table 2. These observations are in line with previous autumn seasons during preconstruction and operational periods of SNWF monitoring, with observed influxes of most soaring birds coinciding with the occurrence of westerly winds.

The species composition of soaring birds in these six documented daily spikes of increased occurrence differed, although all predominantly involved either White storks, Pelicans and/or Common Buzzards. Pelicans were most numerous in 3 of 6 days of intensive migration (22 September, 26 September and 6 October: Fig. 8). White storks dominated in two of the observed days with westerly winds (4 and 22 August). Common buzzards dominated the observed increase at the end of October (30 October).

Figure 9. Correlation between predominant daily wind direction and number of soaring birds observed in SNWF in autumn 2016. Red line shows the linear relationship from which the statistics were derived.

We undertook a simple correlation between daily records of the total number of soaring birds (counts of the 24 soaring species) after a Box-Cox transformation, against the prevailing daily wind direction (Fig. 9). The correlation coefficient (r = 0.4) for the daily prevalence of a westerly wind and the daily count of soaring migrants was statistically significant (p < 0.01). The ‘westerly wind’ metric explained 16% (r2) of the observed daily counts of soaring birds (Fig. 9). In other words significantly higher numbers of ‘soaring’ migrants were associated with days when winds were more westerly and fewer migrants were seen when wind conditions deviated further from the west.

While this result was strongly supportive of the role of the westerly winds in generating the presence of soaring migrants at SNWF, it should be noted that this is against a background of other factors which may militate against such a finding; and so not lead to a very strong relationship. For example, several of the species classed as ‘soaring’ are not entirely dependent on wind conditions for their migration and can, and do, engage in active flight (e.g. falcons Falco spp. and harriers Circus spp.). Also birds’ migration phenology is involved: if no or few birds happen, for other reasons, to be actively engaged in migration on the main flyway to the west of SNWF then there will be few birds that westerly winds would guide eastwards to SNWF. We see this in the results for particular species (described above) when, for example, White storks were not recorded in relatively large numbers on every day when there were westerly winds; but the key finding was that large numbers of White storks were only recorded on days with westerly winds.

28 Despite these factors which may militate against a simple correlative approach illustrating a relationship between westerly winds and numbers of soaring migrants, these analyses from autumn 2016 data confirm previous data analyses from other years, presented in earlier reports (http://www.aesgeoenergy.com/site/Studies.html) indicating that SNWF is situated to the east of the main migratory flyway and so only occasionally hosts major numbers of migrants when -non prevailing- westerly wind conditions shift birds from the flyway. These numbers are consistently lower than stated by BSPB before SNWF was approved for operation.

Turning to collision risk and collision mortality: in all days with intensive bird migration when potentially sensitive species were present (Figure 8 & Table 10) the application of the Turbine Shutdown System (TSS) probably contributed to a reduced risk of collision, and provided a safety mechanism to reduce collision risk for single birds and flocks of endangered or sensitive bird species (Table 10). Documentation of searches for collision victims during autumn 2016 are considered next.

Collision victim monitoring

After two trials for carcass removal and efficiency of the carcass searches in autumn, described in detail in the report for autumn 2014 (http://www.aesgeoenergy.com/site/tcs%20(33).html), a frequency of seven days between searches was defined as optimal to provide objective and cost-effective information about the number of bird collisions with turbines of SNWF.

The numbers of turbines searched during every autumn of operational period of the wind farm are presented in Table 11. The increase of total searches in autumn 2014, 2015 and 2016 was due to the increased monitoring period, until the end of October.

Table 11. Number of carcass searches per autumn and turbine in the operational period of SNWF.

searches Autumn Autumn Autumn Autumn Autumn Autumn Autumn Autumn Turbine Turbine number Total 2010 2011 2012 2013 2014 2015 2016

8 6 8 8 10 13 14 16 75 9 6 8 7 10 12 13 14 70 10 6 7 10 10 14 13 13 73 11 6 7 9 11 17 14 12 76 12 6 10 9 11 19 13 13 81 13 6 9 9 9 17 14 13 77 14 6 9 7 10 15 13 14 74 15 6 9 7 10 15 13 13 73 16 6 6 9 10 15 13 12 71 17 6 6 9 12 13 13 14 73 18 6 4 8 12 14 13 14 71 19 6 8 9 12 15 12 13 75 20 6 9 10 12 14 15 13 79 21 1 6 8 10 16 14 13 68

searches Autumn Autumn Autumn Autumn Autumn Autumn Autumn Autumn Autumn Autumn Autumn Autumn Autumn Turbine Turbine number Total 2010 2011 2012 2013 2014 2015 2016

22 6 6 8 13 14 15 14 76 23 6 6 8 10 18 13 15 76 24 6 7 7 10 16 14 15 75 25 6 2 8 9 16 13 18 72 26 6 8 8 13 13 14 13 75 27 6 2 8 11 14 15 12 68 28 6 2 5 12 13 15 13 66 29 6 8 7 10 16 17 16 80 31 1 9 7 11 15 14 13 70 32 6 9 8 11 15 15 13 77 33 6 8 7 9 18 14 13 75 34 6 8 7 10 15 15 13 74 35 7 8 7 10 15 14 13 74 36 6 9 7 10 13 13 14 72 37 6 9 9 13 15 14 13 79 38 6 9 6 10 14 12 14 71 39 6 8 7 10 16 14 15 76 40 6 7 8 9 16 16 15 77 41 6 7 6 11 18 14 14 76 42 7 7 7 10 15 14 15 75 43 11 9 7 10 15 14 15 81 44 11 7 7 10 15 15 15 80 45 6 8 8 10 13 14 10 69 46 6 9 8 10 14 14 15 76 47 6 9 7 10 15 16 14 77 48 6 9 7 10 14 15 15 76 49 6 10 7 13 14 13 13 76 50 6 10 7 11 15 14 15 78 51 6 9 7 9 14 13 14 72 52 6 9 5 9 15 13 16 73 53 6 9 6 10 13 13 16 73 54 6 8 7 8 15 14 15 73 55 6 9 7 10 18 14 15 79 56 6 8 7 9 14 14 15 73 57 6 9 7 8 14 14 17 75 58 6 9 7 9 14 15 14 74 59 7 9 7 9 16 14 13 75

searches Autumn Autumn Autumn Autumn Autumn Autumn Autumn Autumn Autumn Autumn Autumn Autumn Autumn Turbine Turbine number Total 2010 2011 2012 2013 2014 2015 2016

60 6 9 7 11 15 14 16 78 Total 315 404 389 537 777 725 715 3862

Because of technical maintenance and consequent limited access some turbines were not searched with equal frequency, but as these turbines were not operational in this time period around such maintenance then respective collision risk would be accordingly lower.

Under this search regime during the 2016 autumn migration period, nine sets of remains were found that could be attributed to collision with turbine blades. The number of birds found dead under turbines in 2016 and the species’ conservation status according to the Bulgaria Red Data book and IUCN are presented in Table 12.

Table 12. Collision victims recorded in autumn 2016.

English name Latin name Number of Red Data book IUCN carcasses Magpie Pica pica 1 Not listed Least Concern Eurasian Jay Garrulus glandarius 1 Not listed Least Concern House martin Delichon urbica 1 Not listed Least Concern Spotted flycatcher Muscicapa striata 1 Not listed Least Concern Red Backed shrike Lanius collurio 1 Not listed Least Concern Kestrel Falco tinnunculus 1 Not listed Least Concern Goldcrest Regulus regulus 1 Not listed Least Concern European starling Sturnus vulgaris 1 Not listed Least Concern Yellow-legged Gull Larus michahellis juv. 1 Not listed Least Concern

Table 13. The number of carcasses attributable to collision with wind turbines found during autumn migration between 2010 and 2016 in SNWF. For further details see Methods and reports on the autumn migration period in previous years. Carcasses attributable to Conservation status according Species collision to IUCN (IUCN 3.1) Alauda arvensis 3 Least Concern

Apus apus 3 Least Concern Ardea purpurea 1 Least Concern

Acrocephalus palustris 1 Least Concern Buteo buteo 1 Least Concern Crex crex 1 Least Concern

Delichon urbicum 3 Least Concern Gyps fulvus 1 Least Concern Falco tinnunculus 2 Least Concern Falco vespertinus 1 Near Threatened Hirundo rustica 2 Least Concern Lanius collurio 2 Least Concern Larus ridibundus 1 Least Concern

31 Carcasses attributable to Conservation status according Species collision to IUCN (IUCN 3.1) Larus michahellis 6 Least Concern

Oreolus oreolus 1 Least Concern

Sylvia atricapilla 1 Least Concern Regulus regulus 1 Least Concern Sturnus vulgaris 1 Least Concern Pica pica 1 Least Concern Garrulus glandarius 1 Least Concern Muscicapa striata 1 Least Concern 35

IUCN criteria were used for evaluation of bird conservation status because of the unknown origin of migratory populations in autumn when the movements of birds found dead can cover different continents. National criteria for the same species would be applicable for breeding populations of the same species in the breeding period in spring. The mortality at SNWF for seven autumn seasons of carcass searches, typically under every turbine every week, cannot be remotely considered influential for the populations of any of the affected species.

CONCLUSIONS

Additional data collected in the autumn 2016 by standard methods were consistent with and comparable to previous years’ efforts, and confirmed the previous results and allowed continued evaluation of the long term effect of SNWF on bird migration. The long term monitoring in the same area has allowed the following conclusions:

1. The numbers of species passing through the SNWF territory in autumn varied by year with no trend for a decrease after SNWF was constructed and started its operation (Table 1).

2. The absolute number of observed birds naturally varied by year but with no trend for a decrease after SNWF was constructed and started its operation (Table 2).

3. The altitude of flight varied by years but with no overall trend for an increase after SNWF was constructed and started its operation (Table 4 and Fig. 5).

4. There is no evidence for change in migratory direction (avoidance) associated with the wind farm territory. At a gross scale, birds did not demonstrate macro- avoidance of the turbines that could be considered as a change of migratory direction and, thereby, a change of migratory route (Tables 6, 7, 8 and Fig. 6).

5. The occurrence of autumn migrants in all nine autumn seasons was strongly correlated with typically short periods of a few days when strong westerly winds occurred and deflected birds eastwards from the main migration corridor (Via Pontica) further to the west.

6. During seven years of wind farm operation, carcass searches during the autumn periods revealed a total of 35 collision victims of 21 species of birds.

32 7. Records of collision mortality do not indicate any possibility of an adverse impact of SNWF on any bird population passing through the wind farm territory.

8. The application of the Turbine Shutdown System (TSS) may have made a contribution to the low level of direct mortality registered in the operational period of SNWF for several species identified as being sensitive to collision. Although not formally analysed, micro avoidance of turbine blades also appears to be very high, despite an apparent lack of macro avoidance of the wind farm. Even in the absence of TSS and micro avoidance, however, it is highly unlikely that the pre- construction predictions of mortality would have been observed, in large part because these predictions were based on inflated estimates of the numbers of migrants that “occur” at SNWF.

9. The substantial data collected in seven autumn seasons indicate that the operation of SNWF does not constitute an obstacle or threat, either physically or demographically, to populations of migrants passing through its environs.

REFERENCES

Batschelet E. 1981. Circular Statistics in Biology. Academic Press Inc., New York.

Bibby, C. J., Burgess, N.D. & Hill, D.H. 1992. Bird Census Techniques. London, UK: Academic Press.

Morrison, M. 1998. Avian Risk and Fatality Protocol. Report NREL/SR-500-24997. National Renewable Energy Laboratory. U.S. Department of Energy. 29

RSK Environment Ltd 2008. Saint Nikola Kavarna Wind Farm Supplementary Information Report. RSK Environment report to AES Geo Energy.